SOTAVerified

Zero-Shot Learning

Zero-shot learning (ZSL) is a model's ability to detect classes never seen during training. The condition is that the classes are not known during supervised learning.

Earlier work in zero-shot learning use attributes in a two-step approach to infer unknown classes. In the computer vision context, more recent advances learn mappings from image feature space to semantic space. Other approaches learn non-linear multimodal embeddings. In the modern NLP context, language models can be evaluated on downstream tasks without fine tuning.

Benchmark datasets for zero-shot learning include aPY, AwA, and CUB, among others.

( Image credit: Prototypical Networks for Few shot Learning in PyTorch )

Further readings:

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1601–1650 of 1864 papers

TitleStatusHype
Large Language Models in the Workplace: A Case Study on Prompt Engineering for Job Type Classification—0
Large Model for Small Data: Foundation Model for Cross-Modal RF Human Activity Recognition—0
Large-Scale Zero-Shot Image Classification from Rich and Diverse Textual Descriptions—0
LaSQuE: Improved Zero-Shot Classification from Explanations Through Quantifier Modeling and Curriculum Learning—0
Latent Embeddings for Zero-shot Classification—0
Latent-Variable Generative Models for Data-Efficient Text Classification—0
Learn from Anywhere: Rethinking Generalized Zero-Shot Learning with Limited Supervision—0
Learning Aligned Cross-Modal Representation for Generalized Zero-Shot Classification—0
Learning Attention Propagation for Compositional Zero-Shot Learning—0
Learning Class Prototypes via Structure Alignment for Zero-Shot Recognition—0
Learning Clusterable Visual Features for Zero-Shot Recognition—0
Learning Compositional Representations for Effective Low-Shot Generalization—0
Learning Cross-domain Semantic-Visual Relationships for Transductive Zero-Shot Learning—0
Learning Discriminative Latent Attributes for Zero-Shot Classification—0
Learning Classifiers for Domain Adaptation, Zero and Few-Shot Recognition Based on Learning Latent Semantic Parts—0
Learning from Language Description: Low-shot Named Entity Recognition via Decomposed Framework—0
Learning Graph-Based Priors for Generalized Zero-Shot Learning—0
Learning Joint Feature Adaptation for Zero-Shot Recognition—0
Learning Primitive Relations for Compositional Zero-Shot Learning—0
Learning Robust Visual-semantic Mapping for Zero-shot Learning—0
Learning shared manifold representation of images and attributes for generalized zero-shot learning—0
Learning Simplifications for Specific Target Audiences—0
Learning Structured Semantic Embeddings for Visual Recognition—0
Learning the Redundancy-free Features for Generalized Zero-Shot Object Recognition—0
Learning to Classify Events from Human Needs Category Descriptions—0
Learning to Name Classes for Vision and Language Models—0
Learning to Recognize the Unseen Visual Predicates—0
Learning the Compositional Spaces for Generalized Zero-shot Learning—0
Learning Unseen Emotions from Gestures via Semantically-Conditioned Zero-Shot Perception with Adversarial Autoencoders—0
Transferring Knowledge from Text to Video: Zero-Shot Anticipation for Procedural Actions—0
Learning Visual Emotion Representations From Web Data—0
Learning Visually Consistent Label Embeddings for Zero-Shot Learning—0
Semantic-Guided Multi-Attention Localization for Zero-Shot Learning—0
Learning without Seeing nor Knowing: Towards Open Zero-Shot Learning—0
Learn to Adapt for Generalized Zero-Shot Text Classification—0
Less is more: zero-shot learning from online textual documents with noise suppression—0
LETS-GZSL: A Latent Embedding Model for Time Series Generalized Zero Shot Learning—0
Leverage Weakly Annotation to Pixel-wise Annotation via Zero-shot Segment Anything Model for Molecular-empowered Learning—0
Leveraging Contextual Relatedness to Identify Suicide Documentation in Clinical Notes through Zero Shot Learning—0
Leveraging Knowledge Graphs for Zero-Shot Object-agnostic State Classification—0
Wisdom of Instruction-Tuned Language Model Crowds. Exploring Model Label Variation—0
Leveraging Large Language Models to Extract Information on Substance Use Disorder Severity from Clinical Notes: A Zero-shot Learning Approach—0
Leveraging Large-Scale Pretrained Vision Foundation Models for Label-Efficient 3D Point Cloud Segmentation—0
Leveraging MoE-based Large Language Model for Zero-Shot Multi-Task Semantic Communication—0
Leveraging Semantic Embeddings for Safety-Critical Applications—0
Leveraging Type Descriptions for Zero-shot Named Entity Recognition and Classification—0
Leveraging Vision-Language Embeddings for Zero-Shot Learning in Histopathology Images—0
Leveraging Weakly Annotated Data for Hate Speech Detection in Code-Mixed Hinglish: A Feasibility-Driven Transfer Learning Approach with Large Language Models—0
Light Field Super-Resolution With Zero-Shot Learning—0
Linear Ranking Analysis—0
Show:102550
← PrevPage 33 of 38Next →
All datasetsCUB-200-2011MedConceptsQASUN AttributeAwA2Caltech-101CIFAR-10CIFAR-100COCO-MLTDTDFGVC-AircraftFlowers-102Food-101

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy87.5—Unverified
2DUETaverage top-1 classification accuracy72.3—Unverified
3Composeraverage top-1 classification accuracy69.4—Unverified
4HDC-ZSC-MLPaverage top-1 classification accuracy65.6—Unverified
5ZSL_TF-VAEGANaverage top-1 classification accuracy64.9—Unverified
6ZLaPAccuracy64.3—Unverified
7ZLaP*Accuracy64.2—Unverified
8HDC-ZSCaverage top-1 classification accuracy63.8—Unverified
9SPOTaverage top-1 classification accuracy62.9—Unverified
10f-VAEGAN-D2average top-1 classification accuracy61—Unverified
#ModelMetricClaimedVerifiedStatus
1dmis-lab/biobert-v1.1Accuracy26.15—Unverified
2meta-llama/Meta-Llama-3-8B-InstructAccuracy25.84—Unverified
3epfl-llm/meditron-7bAccuracy25.75—Unverified
4dmis-lab/meerkat-7b-v1.0Accuracy25.68—Unverified
5meta-llama/Meta-Llama-3-8B-InstructAccuracy25.65—Unverified
6HuggingFaceH4/zephyr-7b-betaAccuracy25.54—Unverified
7dmis-lab/biobert-v1.1Accuracy25.46—Unverified
8epfl-llm/meditron-70bAccuracy25.36—Unverified
9epfl-llm/meditron-70bAccuracy25.26—Unverified
10HuggingFaceH4/zephyr-7b-betaAccuracy25.06—Unverified
#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy77.3—Unverified
2SPOT (VAEGAN)average top-1 classification accuracy66.04—Unverified
3ZSL_TF-VAEGANaverage top-1 classification accuracy66—Unverified
4f-VAEGANaverage top-1 classification accuracy64.7—Unverified
5DUET (Ours)average top-1 classification accuracy64.4—Unverified
6LisGANaverage top-1 classification accuracy61.7—Unverified
7TCNaverage top-1 classification accuracy61.5—Unverified
8f-CLSWGANaverage top-1 classification accuracy60.8—Unverified
9Cycle-WGANaverage top-1 classification accuracy59.9—Unverified
#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy86.4—Unverified
2ZSL-KGaverage top-1 classification accuracy78.08—Unverified
3ZSL_TF-VAEGANaverage top-1 classification accuracy72.2—Unverified
4DUET (Ours)average top-1 classification accuracy69.9—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy84—Unverified
2ZLaP*Accuracy83.1—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy93.6—Unverified
2ZLaPAccuracy93.4—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy74.2—Unverified
2ZLaPAccuracy74—Unverified
#ModelMetricClaimedVerifiedStatus
1ViT-B/16Average mAP60.17—Unverified
2ResNet-50Average mAP56.19—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy51.2—Unverified
2ZLaP*Accuracy51—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy29.1—Unverified
2ZLaP*Accuracy29—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy75.9—Unverified
2ZLaP*Accuracy75.5—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy87.9—Unverified
2ZLaPAccuracy87.8—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPTop 1 Accuracy72.1—Unverified
2ZLaP*Top 1 Accuracy72.1—Unverified
#ModelMetricClaimedVerifiedStatus
1HiTeAAccuracy21.7—Unverified
2HiTeAAccuracy0.46—Unverified
#ModelMetricClaimedVerifiedStatus
1HiTeAAccuracy37.4—Unverified
2HiTeAAccuracy0.56—Unverified
#ModelMetricClaimedVerifiedStatus
1SPOTaverage top-1 classification accuracy71.9—Unverified
2ZSL_TF-VAEGANaverage top-1 classification accuracy70.8—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy90—Unverified
2ZLaP*Accuracy89—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy71.8—Unverified
2ZLaPAccuracy71.2—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy71.4—Unverified
2ZLaPAccuracy71—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy76.3—Unverified
2ZLaPAccuracy76.3—Unverified
#ModelMetricClaimedVerifiedStatus
1CLIP(ViT-B/16)Average mAP85.77—Unverified
2CLIP(ResNet-50)Average mAP84.3—Unverified
#ModelMetricClaimedVerifiedStatus
1ZSL-KGTop-160.54—Unverified
#ModelMetricClaimedVerifiedStatus
1zsl_ADAAverage Per-Class Accuracy70.9—Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy63.2—Unverified
#ModelMetricClaimedVerifiedStatus
1MSDAPearson correlation coefficient (PCC)0.52—Unverified
#ModelMetricClaimedVerifiedStatus
1SeViLAAccuracy72.3—Unverified
#ModelMetricClaimedVerifiedStatus
1M^2-EncoderAccuracy80.7—Unverified
#ModelMetricClaimedVerifiedStatus
1FrozenBiLMAccuracy51.5—Unverified
#ModelMetricClaimedVerifiedStatus
1CZSLA-acc36—Unverified
#ModelMetricClaimedVerifiedStatus
1ZS3Netk=10 mIOU26.3—Unverified
#ModelMetricClaimedVerifiedStatus
1ZSL-KGAccuracy88.98—Unverified
#ModelMetricClaimedVerifiedStatus
1VideoChat2Accuracy40.6—Unverified